Defined sequence RNA synthesis in the 3′→5′ direction is now well established and currently in use for synthesis and development of a vast variety of therapeutic grade RNA aptamers, tRNA's, siRNA and biologically active RNA molecules. This approach utilizes a ribonucleoside with suitable N-protecting group: generally 5′-Protecting group, the most popular being dimethoxytriphenyl, i.e. the DMT group; 2′-protecting group, out of which most popular is t-Butyldimethylsilyl ether; and, a 3′-phosphoramidite, the most popular of which is cyanoethyl diisopropyl (component 1). This component is then coupled with a nucleoside with a suitable N-protecting group, 2′ or 3′ succinate of a ribonucleoside attached to a solid support (component 2). The coupling of component 1 and 5′-OH-n-protected-2′,3′-protected-nucleoside (component 3) are also achieved in solution phase in presence of an activator to lead to dimers and oligoribonucleotides, followed by oxidation (3′→5′ direction synthesis), also lead to protected dinucleoside having a 3′-5′-internucleotide linkage, Ogilvie, K. K., Can. J. Chem., 58, 2686, 1980 (scheme 1).
However, the synthesis of RNA in the reverse direction (5′-3′ direction) has not been achieved so far to the best of our knowledge.
The 2′-silyl ethers of component 1 have been developed extensively and they are known to have remarkable stability. Solvolysis of silyl ethers have been extensively studied and it is known that bulky alkyl silyl ethers have a high degree of stability; Bazani, B and Chvalowski, V Chemistry of Organosilicon compounds, Vol. 1, Academic Press, New York, 1965. Extensive research work was subsequently done by Ogilvie and coworkers as 2′-hydroxy protecting group for oligo ribonucleotide synthesis (Ogilvie, K. K., Sadana, K. L, Thompson, E. A., Quilliam, M. A., and Westmore, J. B Tetrahedron Letters, 15, 2861-2864, 1974; Ogilvie, K. K., Beaucage, S. L, Entwistle, D. W., Thompson, E. A., Quilliam, M. A., and Westmore, J. B. J. Carbohydrate Nucleosides Nucleotides, 3, 197-227, 1976; Ogilvie, K. K. Proceedings of the 5th International Round Table on Nucleosides, Nucleotides and Their Biological Applications, Rideout, J. L., Henry, D. W., and Beacham L. M., III, eds., Academic, London, pp. 209-256, 1983).
These studies subsequently led to continued developments of methods which were amenable to both solution and solid phase oligonucleotide synthesis, and the first chemical synthesis of RNA molecules of the size and character of tRNA (Usman, N., Ogilvie, K. K., Jiang, M.-Y.,and Cedergren, R. J. J. Am. Chem. Soc. 109, 7845-7854, 1987; Ogilvie, K. K., Usman, N., Nicoghosian, K, and Cedergren, R. J. Proc. Natl. Acad Sci. USA, 85, 5764-5768, 1988; Bratty, J., Wu, T., Nicoghosian, K., Ogilvie, K. K., Perrault, J.-P., Keith, G. and Cedergren, R., FEBS Lett. 269, 60-64, 1990). The literature has been amply reviewed in subsequent excellent publication: Gait, M. J., Pritchard, C. and Slim, G., Oligonucleotides and Their Analogs: A Practical Approach (Gait, M. J., ed.), Oxford University Press Oxford, England, pp 25-48, 1991. Other protecting groups which have been lately employed for RNA synthesis are: bis(2-acetoxyethyl-oxy)methyl (ACE), Scaringe, S. A., Wincott, F. E., Caruthers, M. H., J. Am. Chem. Soc., 120: 11820-11821, 1998; triisopropylsilyloxy methyl (TOM), Pitsch, S., Weiss, P. A., Jenny, L., Stutz, A., Wu, X., Hely. Chim. Acta. 84, 3773-3795, 2001 and t-butyldithiomethyl (DTM) (structure 1), Semenyuk, A., Foldesi, A., Johansson, T., Estmer-Nilsson, C., Blomgren, P., Brannvall, M., Kirsebom, L. A., Kwiatkowski, M., J. Am. Chem. Soc., 128: 12356-12357, 2006 have been introduced. However, none of these processes is amenable to carry out the synthesis of RNA in reverse direction (5′→3′direction); hence they lack the capability of the convenient and efficient introduction of many ligands and chromophores at the 3′-end of RNA molecules, achievable through reverse direction synthesis.
Chemically modified RNA have been synthesized having modified arabino sugars, 2′-deoxy-2′-fluoro-beta-D_arabinonucleic acid (FANA; structure 2)) and 2′-deoxy-4′-thio-2′-fluoro-beta-D_arabinonucleic acid (4′-Thio-FANA; structure 3) into sequences for siRNA activities, Dowler, T., Bergeron, D., Tedeschi, Anna-Lisa, Paquet, L., Ferrari, N., Damha, M. J. Nucl. Acids Res., 34, 1669-1675, 2006. Amongst the several new 2′-protecting groups the chemistry for which have been developed, the 2′-protecting 2-cyanoethoxymethyl (CEM) (structure 4) has been shown for producing very long RNA, however, which also carries out RNA synthesis in the conventional, i.e., the 3′→5′direction. Furthermore, the quality of RNA produced by these processes remains in question.

The chemical synthesis of RNA is desirable because it avoids the inefficiencies and limitation of scale of synthesis such as by in vitro transcription by T7 RNA polymerase, Helm, M., Brule, H., Giege, R., Florence, C., RNA, 5:618-621, 1999. Chemical synthesis of RNA is desirable for studies of RNA structure and function, and many useful modifications can be achieved selectively, such as site specific introduction of functional groups; viz., disulphide cross linking as a probe of RNA tertiary structures, Maglott, E. J., Glick, G. D., Nucl. Acids Res., 26: 1301-1308, 1999.
Synthesis of long RNA is very important for biologically active molecules such as tRNA, and such synthesis has been achieved; Persson, T., Kutzke, U., Busch, S., Held, R., Harmann, R. K., Bioorgan. Med. Chem., 9:51-56, 2001; Ogivie, K. K., Usman, N., Nicoghosian, K., Cedrgren, R. J., Proc. Natl. Acad. Sci., USA, 85:5764-5768, 1988; Bratty, J., Wu, T., Nicoghosian, K., Ogilvie, K. K., Perreault, J.-P., Keith, G., Cedergren, R. J., F.E.B.S. Lett., 269:60-64, 1990; Gasparutto, D., Livache, T., Bazin, H., Duplaa, A. M., Guy, A., Khorlin, A., Molko, D., Roget, A., Teoule, R., Nucl. Acids. Res., 20:5159-5166, 1992; Goodwin, J. T., Stanick, W. A., Glick, G. D., J. Org. Chem., 59:7941-7943, 1994. None of the techniques mentioned in this paragraph, however contemplate the synthesis of RNA in reverse direction (5′→3′direction), and hence the practical and convenient introduction of a number of groups required for selective introduction at 3′-end remains elusive.
The present inventors have observed higher coupling efficiency per step during automated oligo synthesis with our reverse RNA amidites, resulting in a greater ability to achieve higher purity and produce very long oligos. They also demonstrate that the process of this invention leads to oligonucleotides free of M+1 species, which species lead to closer impurities as shoulder of desired peak during HPLC analysis or purification or Gel purification.
The t-butyldimethyl silyl protecting group on 2′-hydroxyl of ribonucleosides has been the group of choice for making 3′-phosphoramidites and for utilizing them for oligonucleotide synthesis which have been shown to migrate to 3′-hydroxyl position rather easily. This has been documented amply and in detail (Ogilvie, K. K., and Entwistle, D. W. Carbohydrate Res., 89, 203-210, 1981; Wu, T., and Ogilvie, K. K. J. Org. Chem., 55, 4717-4734, 1990). Such migration complicates the synthesis of the desired phosphoramidites and requires an efficient method of purification that clearly resolves corresponding isomers and prevents any contamination of the final monomer.
The present invention is directed towards the synthesis of high purity RNA's, specifically to introduce selected groups at 3′-end of oligonucleotides of synthetic RNA's. Such RNA's have vast application in therapeutics, diagnostics, drug design and selective inhibition of an RNA sequence within cellular environment, blocking a function of different types of RNA present inside cell.
Silencing gene expression at mRNA level with nucleic acid based molecules is a fascinating approach. Among these RNA interference (RNAi) has become a proven approach which offers great potential for selective gene inhibition and shows great promise for application in the control and management of various biochemical and pharmacological processes. Early studies by Fire et al., Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E., and Mello, C. C, Nature, 391, 806-811, 1998, showed that RNA interference in Caenorhabditis elegans is mediated by 21 and 22 nucleotide RNA sequences. This was further confirmed as a general phenomenon of specific inhibition of gene expression by small double stranded RNA's being mediated by 21 and 22 nucleotide RNA's, Genes Dev., 15, 188-200, 2001. Simultaneous studies by Capie, N. J., Parrish, S., Imani, F., Fire, A., and Morgan, R. A., confirmed such phenomenon of specific gene expression by small double stranded (dS) RNAs in invertebrates and vertebrates alike. Subsequently a vast amount of research led to the confirmation of above studies and established RNAi as a powerful tool for selective, and very specific gene inhibition and regulation; Nishikura, K., Cell, 107, 415-418, 2001; Nykanen, A., Haley, B., Zamore, P. D., Cell, 107, 309-321, 2001; Tuschl, T., Nat. Biotechnol., 20, 446-448, 2002; Mittal, V., Nature Rev., 5, 355-365, 2004; Proc. Natl. Acad. Sci. USA, 99, 6047-6052, 2002; Donze, O. & Picard, D., Nucl. Acids. Res., 30, e46,2002; Sui, G., Soohoo, C., Affar el, B., Gay, F., Shi, Y., Forrester, W.c., and Shi, Y., Proc. Natl. Acad. Sci. USA, 99, 5515-5520, 2002; Paddison, P. J., Caudy, A. A., Bernstein, E., Hannon, G. J.; and Conklin, D. S., Genes Dev., 16, 948-959, 2002.
Besides the natural double stranded (ds) RNA sequences, chemically modified RNA have been shown to cause similar or enhanced RNA interference in mammalian cells using 2′-deoxy-2′-fluoro-beta-D_arabinonucleic acid (FANA) into sequences for siRNA activities, Dowler, T., Bergeron, D., Tedeschi, Anna-Lisa, Paquet, L., Ferrari, N., Damha, M. J., Nucl. Acids Res., 34, 1669-1675, 2006.
Various other modifications to improve siRNA properties have been pursued, which include alteration in backbone chemistry, 2′-sugar modifications, and nucleobase modifications, some of which have been recently reviewed; Nawrot, B, and Sipa, K., Curr. Top. Med. Chem., 6,913-925, 2006; Manoharan, M. Curr. Opin. Chem. Biol., 8, 570-579, 2004. The PS modifications of siRNA are well tolerated, although some reports indicate increased toxicity and somewhat reduced efficacy; Harborth, J., Elbasir, S. M., Vandenburgh, K., Manninga, H., Scaringe, S. A., Weber, K., Tuschl, T., Antisense Nucleic Acid Drug Dev., 13, 83-105, 2003. Among these is also the 2′-Omethyl modification, although it maintains A form (RNA like) helix, and has been shown to be either retaining or reducing siRNA activity depending on the number of such modifications within a sequence, Chiu, Y. L., Rana, T. M., RNA, 9, 1034-1048, 2003. It has also been shown that extensive 2′-O-Methyl modification of a sequence can be made in the sense strand without loss of siRNA activity, Kraynack, B. A., Baker, B. F., RNA, 12, 163-176, 2006.
Bicyclic locked nucleic acids (LNA's) that confer high binding affinity have also been introduced in siRNA sequences, especially when the central region of siRNA sequence is avoided, Braash, D. A., Jensen, S., Liu, Y., Kaur, K., Arar, K., White, M. A., and Corey, D. R., Biochemistry, 42, 7967-7995, 2003. Similarly altritol sugar modified oligonucleotides (ANA) have recently been reported. Altrilol sugar offers a rigid conformation and is shown to form very stable duplexes with RNA in a sequence specific manner, and further shown to stay in A (RNA type) conformation. It was shown that ANA modified siRNAs targeting MDR1 gene exhibited improved efficacy as compared to unmodified controls, specifically effective when this modification was near the 3′-end of sense or antisense strand; Fisher, M., Abramov, M., Aerschot, A. V., Xu, D., Juliano, R. L., Herdewijn, P., Nucl. Acids Res., 35, 1064-1074, 2007.
Among the various requirements for effective RNA interference (RNAi) to take place, a number of observations and facts have been established. Thus the RNA has to be double stranded (ds)) in the region of identity to the target. For the chemical requirement, besides the capability of 5′-end to be converted to triphosphates, modifications of A, C, G were found to be fully compatible with interference activity. Backbone modification of RNA, such as 2′-fluoro, 2′-amino uracil, 2′-deoxy thymidine, and 2′-deoxycytidine appear to stabilize the modified RNA in the resulting double strand. Amongst the nucleoside base modification 5-bromouracil, 4-thiouracil, 5-iodouracil, 5-(3-aminoallyl)-uracil, inosine are readily incorporated in the RNA intereference complex (RNAi & RISC complex). Similarly inosine may be substituted for guanosine.
It has been shown that cholesterol-conjugated siRNA can achieve delivery into cells and silence gene expression. Further, it has been shown that lipid conjugated siRNA, bile acids, and long chain fatty acids can mediate siRNA uptake into cells and silence gene expression in vivo. Efficient and selective uptake of siRNA conjugates in tissues is dependent on the maximum association with lipoprotein particles, lipoprotein/receptor interactions and transmembrane protein mediated uptake. It has been shown that high density lipoproteins direct the delivery of of siRNA into liver, gut, kidney and steroidal containing organs. It has been further shown that LDL directs siRNA primarily to the liver and that LDL receptor is involved in the delivery of siRNA. These results show great promise for siRNA uptake by an appropriate delivery system which can be exploited in development of therapeutics. (Article by Marcus Stoffel, OTS, 2007, p. 64.)
It has been proposed that siRNA can be designed with chemical modifications to protect against nuclease degradation, abrogate inflammation, reduce off target gene silencing, and thereby improve effectiveness for target genes. Delivery vehicles or conjugates of lipids and other lipophilic miolecules which allow enhanced cellular uptake are essential for therapeutic developments. Such siRNA's are presently being developed for human target validation, interfering with diseases pathways and developing new frontier for drug development. Alan Sachs, Merck, Oligonucleotide Therapeutics Conference (OTS), page 80, 2007.
The 3′-end of sense strand of siRNA can be modified and has been shown to tolerate modification, and that attachment of ligands is most suited at this end (FIGS. 1 & 2), as detailed in a number of key publications; siRNA function in RNAi: a chemical modification, Ya-Lin Chiu and Tariq Rana, RNA, 9, 1034-1048, 2003; M. Manoharan, Curr. Opin. Chem. Biol, 6, 570-579, 2004; Nawrot, B. and Sipa, K., Curr. Top. Med. Chem., 6, 913-925, 2006; Scaringe, S., Marshall, W. S., Khvorova, A., Naty. Biotechnol., 22, 326-30, 2004.

The introduction of lipophilic or hydrophobic groups and enhancing of siRNA delivery and optimization of targets has been addressed and achieved through bio-conjugation (Diagram 1). Generally the attachment is done, preferably at the 3′-end of senses strand, and occasionally on the 3′-end of the antisense strand. The design of nuclease resistant siRNA has been the subject of intense research and development recently in order to develop effective therapeutics. Thus base modifications such as, 2-thiouridine, pseudouridine, dihydrouridine have revealed the effect on conformations of RNA molecules and the associated biological activity; Sipa, K., Sochacka, E., Kazmierczak-Baranska, J., Maszewska, M., Janicka, M., Nowak, G., Nawrot, B., RNA, 13, 1301-1316, 2007. It was shown that 2′-modified RNA's especially 2′-Fluoro have great resistance towards nuclease and are biological active in-vivo, Layzer, J. M., McCaffrey, A. P., Tanner, A. K., Huang, Z., Kay, M. A., and Sullenger, B. A., RNA, 10, 766-771, 2004. 2′-O-Alkyl-modification, such as 2′-Omethyl's and 2′-O-MOE, Prakash, S., Allerson, C V. R., Dande, P., Vickers, T. A., Siofi, T. A., Jarres, R., Baker, B. F., Swayze, E. E., Griffey, R. H., and Bhat, B., J. Med. Chem., 48, 4247, 4253, 2005. The same authors used 4′-thio modified sugar nucleosides in combination of 2′-O alkyl modification for improving siRNA properties and RNAi enhancement, Dande, P., Prakas, T. P., Sioufi, N., Gaus, H., Jarres, R., Berdeja, A., Swayne, E. E., Griffey, R. H., Bhat, B. K, J. Med. Chem., 49, 1624-1634, 2006. The Replacement of internucleotide phosphate with phosphorothioate and boranophosphates of siRNAs show promise in-vivo, Li, Z. Y., Mao, H., Kallick, D. A., and Gorenstein, D. G., Biochem. Biophys. Res. Comm., 329, 1026-1030, 2005; Hall, A. H. S., Wan, J., Shaughnessy, E. E., Ramsay Shaw, B., Alexander, K. A., Nucl. Acids Res., 32, 5991-6000, 2004.
Bioconjugation of siRNA molecules, biologically RNA molecules, aptamers and synthetic DMNA molecules require, in addition to in vivo stability and appropriate modification of nucleosides, a key feature for cell membrane permeability: Insufficient cross-membrane cellular uptake limits the utility of siRNA's, other single stranded RNA's or even various DNA molecules. Thus cholesterol attached at 3′-end of siRNA has been shown to improve in-vivo cell trafficking and therapeutic silencing of gene, Soutschek, J., Akine, A., Bramlage, B., Charisse, K., Constein, R., Donoghue, M., Elbasir, S., Geickk, A., Hadwiger, P., Harborth, J., Nature, 432, 173-0178, 2004.
Among the various conjugations, besides cholesterol, which have been developed are:                (a) Natural and synthetic protein transduction domains (PTDs), also called cell permeating peptides (CPPs) or membrane permeant peptides (MPPs) which are short amino acid sequences that are able to interact with the plasma membrane. The uptake of MPP-siRNA conjugates takes place rapidly. Such peptides can be conjugated preferably to the 3′-of stand strand.        (b) Other polycationic molecules can be conjugated at the 3′-end of either sense or antisense strand of RNA.        (c) PEG (polyethylene glycols-oligonucleotide conjuagates) have been used in various complex possess significant gene silencing effect after uptake in target cells, Oishi, M., Nagasaki, Y., Itaka, K., Nishiyama, N., and Kataoka, K., J. Am. Chem. Soc., 127, 1624-1625, 2005.        (d) Aptamers have been used for site specific delivery of siRNA's. Since        
Aptamers have high affinity for their targets, the conjugates with siRNA act as excellent delivery system, which result in efficient inhibition of the target gene expression, Chu, T. C., Twu, K. Y., Ellington, A. D. and Levy, M., Nucl. Acids Res., 34(10), e73, 2006. These molecues can once again be conjugated at the 3′-end of siRNA or other biologically active oligonucleotides.                (e) Various lipid conjugations at the 3′-end can be achieved through the present invention and can be utilized for efficient internalization of oligonucleotides. The lipophilic moiety can consist of a hydroxyl function to synthesize a phopsphoramidite. Similarly the lipophilic moiety can have carboxylic function at the terminus. The later can be coupled to a 3′-amino group having a spacer, synthesized by last addition of amino linkers such as C-6 amino linker amidite, of the reverse synthesized oligonucleotide, to the carboxylic moiety using DCC (dicyclohexyl cabodiimide) or similar coupling reagent.        
This research has been reviewed elegantly by Paula, De. D., Bentley, M. V. L. B., Mahao, R. L., RNA, 13, 431-456, 2007.
Another class of RNA, closely related to siRNA are microRNA, commonly referred as miRNA. These are a large class of non coding RNA's which have a big role in gene regulation, Bartel, D. P. Cell, 116,281-297, 2004; He, L., Hannon, G. J. Nat. Rev. Genet, 5:522-531, 2004; Lagos-Quintana, M., Rauhut, R., Lendeckel, W., Tuschl, T., Science, 204:853-858,2001. In human genome there are at least 1000 miRNA scattered across the entire genome. A number of these micro RNA's down regulate large number of target mRNAs, Lim, L. P., Lau, N. C., Garrett-Engele, P., Grimson, A., Schelter, J. M., Castle, J., Bartel, D. P., Linsey, P. S., Johnson, J. M., Nature, 433:769-773, 2005. Different combination of miRNAs are possibly involved in the regulation of target gene in mammalian cell. It has also been shown that siRNA can function as miRNAs; Krek, A., Grun, D., Poy, M. N., Wolf, R., Rosenberg, L., Epstein, E. J., MacMenamin, P., da Piedade, I., Gunsalus, K. C., Stoffel, M., Nat. Genet., 37: 495-500, 2005; Doench, J. G., Petersen, C. P., Sharp, P. A., Genes Dev., 17:438-442, 2003. The miRNA have great potential in therapeutics and in gene regulation, Hammond, S. M., Trends Mol. Med. 12:99-101, 2006. A vast amount of effort is being currently devoted towards understanding miRNA pathways, their role in development and disease, focusing specially on cancer. The miRNA targets are being developed for therapeutic and diagnostics development. A great number of miRNA are being identified and their role is being determined through microarrays, PCR and informatics. Syntheses of RNA designed to target miRNA also require RNA synthesis and other modification as required for siRNA's, for the stability of RNA and the bioconjugation for better cellular uptakes. The reverse synthesis envisioned by the present invention can greatly accelerate the pace of this research and development.
Synthesis of vast variety of therapeutic grade RNA and siRNA requires a modification or labeling of 3′-end of an oligonucleotide. In the case of siRNA, generally it is the 3′-end of sense strand. The synthesis of 3′-end modified RNA requiring lipophilic, long chain ligands or chromophores, using 3′→5′ synthesis methodology is challenging, requires corresponding solid support and generally results in low coupling efficiency and lower purity of the final oligonucleotide, in general, because of the large amount of truncated sequences containing desired hydrophobic modification.
The present inventors have approached this problem by developing reverse RNA monomer phosphoramidites for RNA synthesis in 5′→3′-direction. This arrangement leads to very clean oligonucleotide synthesis allowing for introduction of various modifications at the 3′-end cleanly and efficiently. In order to enhance stability and eliciting of additional favorable biochemical properties, this technique can utilize 2′-5′-linked DNA (Structure 5) and RNA (Structure 6) which have been developed in the past.

For the efficient delivery of RNA and to increase cellular concentration of oligonucleotides, lipids containing oligonucleotides are generally synthesized, since lipids containing synthetic nucleosides enhance uptake of many synthetic nucleoside drugs, like AZT. Lipipoid nucleic acids are expected to reduce the hydrophilicity of oligonucleotides. Similarly hydrophobic molecules such as cholesterol can bind to LDL particles and lipoproteins, and activate a delivery process involving these proteins to transport oligonucleotides. It has also been shown that lipidoic nucleic acids improve the efficacy of oligonucleotides. Shea, R. G., Marsters, J. C., Bischofberger, N., Nucleic Acids Res., 18, 3777, 1990; Letsinger, R. L., Zhang, G., Sun, D. K., Ikeuchi, T., Sarin, P. S., Proc. Natl. Acad. Sci. USA 86, 6553, 1989; Oberhauser, B., and Wagner, E., Nucleic Acids Res., 20, 533, 1992; Saison,-Behmoaras, T., Tocque, B., Rey, I., Chassignol, M., Thuong, N. T., Helene, C. The EMBO Journal, 10, 1111, 1991; Reed, M. W., Adams, A. D., Nelson, J. S., Meye R. B., Jr., Bioconjugate Chem., 2, 217, 1991; Polushin, N. N., Cohen, J., J. Nucleic Acids Res., 22, 5492, 1994; Vu, H., Murphy, M., Riegel, Joyce, M., Jayaraman, K., Nucleosides & Nucleotides, 12, 853, 1993; Marasco, Jr., Angelino, N. J., Paul, B., Dolnick, B. J., Tetrahedron Lett., 35, 3029, 1994. In the studies, the Tm of a series of hydrophobic groups, such as adamantane (structure 7), eicosenoic acid (structure 8), and cholesterol were attached to oligodeoxy nucleotide sequences at the 3′-end and were hybridized to complemetary RNA sequences; the Tm was found to be unaffected, which indicates that such groups do not interefere with oligo hybridization properties; Manoharan, M., Tivel, K. L., Andrade, L. K., Cook, P. D., Tetrahedron Lett., 36, 1995; Manoharan, M., Tivel, K. L., Cook, P. D., Tetrahedron Lett., 36, 3651-3654, 1995; Gerlt, J. A. Nucleases, 2 nd Edition, Linn, S. M., Lloyd, R. S., Roberts, R. J., Eds. Cold Spring Harbor Laboratory Press, p-10, 1993.

Besides the preceeding lipidoic molecules other class of molecules, which have shown high promise are short and long chain polyethyene glycols (PEG), Bonora, G. M., Burcovich, B., Veronese, F. M., Plyasunova, O., Pokrovsky, A. and Zarytova, V., XII International Round Table Conference, Nucleosides, Nucleotides and their Biological Applications, La Jolla, Calif., September, 15-19, PPI 89, 1996. For efficient delivery of synthetic RNA, and DNA molecules PEG attachment to various oligonucleotides have shown to very favorable properties. PEG-oligomers have shown nice enzymatic stability by preventing fast digestion. The thermal melting behaviour was not affected, thereby still retaining properties of double strand formation.
The approach of the present inventors to produce reversed RNA phosphoramidites required selective introduction of 5′-esters, such as 5′-benzouyl, acetyl, levulinyl or substituted-5′-benzoyl-n-protected ribonucleoside. Subsequent introduction of an appropriate protecting group at 2′-position, such as 2′-tBDsilyl or 2′-TOM (triisopropyl silyl methoxy; TOM) was required. In their scheme they show synthesis of selected molecules to achieve this purpose.
In order to produce target compounds, structures (16a-e), the key intermediate required was 2′-silylether-5′-O acylated-N-protected ribonucleoside, compounds (23a-e). The first intermediate required for this was 5′-acylated-N-protected ribonucleoside compound (22a-e). To the best of our knowledge, the compounds (21a-e) have not been reported. The compounds, which are reported for RNA synthesis in the past utilized intermediates, structure 11, 12 and 13. Various 2′ and 3′ acetates with 5′-acetate; viz., 5′-benzoyl protected nucleosides; Reese, B. E., Jarman, M., Reese, C. B., Tetrahedron, 24, 639, 1968; Neilson, T., Werstiuk, E. S., Can. J. Chem. 49, 493, 1971; Neilon, T., Wastrodowski, E. V., Werstiuk, E. S., Can. J. Chem., 51, 1068, 1973; Eckstein, F., Cramer, F., Chem. Ber., 98, 995, 1965; Zemlicka, J., Chladek, S., Tet. Lett., 3057, 1965, Amarnath, V. & Broom, A. D., Chemical Reviews, 77, 183-219,1977. The present invention, however, required free 2′ and 3′ hydroxyl groups, such as in structures (22a-e).

In the present invention, comparative synthesis and purification of RNA's were carried out, both by conventional method (3′→5′) and reverse Direction (5′→3′). Observed were: High Purity of RNAs, Smooth 3′-Conjugation-Cholesterol, HEG and PEG (Polyethylene glycols) & Demonstration of Absence of M+1 in Reverse RNA Synthesis
The details of synthesis scheme are outlined in the Scheme (2). The 2′,3′-Isopropylidene function is utilized to protect the 2′,3′ hydroxyl groups of ribose of n-protected ribonucleosides. A number of preferred n-protecting groups are shown in the Scheme (2). The 5′-hydroxyl group is subsequently protected, preferably with benzoyl group to obtain compounds of general structure 21. The isopropylidene group is then selectively removed under mild acidic conditions well known in the art. This step leads to compounds of general formula 22. Subsequent reaction with TBDM Silyl chloride (tert-butyl dimethyl silyl chloride) leads to mono silyl compound of general formula 23. The present inventors have observed that 3′-TBDMS group, i.e., the formation of compound structure 24 is not preferred in this process. In most of the cases they observed a clean product whose structure was confirmed to be 23 by chemical and analytical methods.
Purification is carried out at each step either via crystallization or column chromatography at each of the steps of the process mentioned above. Subsequent reaction with dimethoxytrityl chloride (DMT-chloride) in pyridine leads to 3′-DMT-2′-TBDMS-n-protected nucleosides of the general structure 26. Each of the compounds were fully purified by column chromatography.
Altough they utilized TBDMS group to produce 2′-TBDMS ether, other silyl ethers can be utilized at this step. A careful aqueous/methanolic NaOH hydrolysis resulted in compounds with free 5′-hydroxyl group, general structure 27.
Selective 5′-benzoyl removal with aqueous or methanolic base is well known in the art. The compounds 3′-DMT-2′-TBDMS-n-protected nucleosides (structure 27) were purified by silica gel column chromatography. The purified compounds (structure 27) were subsequently phosphorylated with phosphorylating reagents, such as n,n-diisopropylamino cyanoethyl phosphonamidic chloride or 2-cyanoethyl,n,n,n,n-tetraisopropyl phosphane to yield the corresponding phosphramidites (structures 16). Both the phosphorylating reagents, n,n-diisopropylamino cyanoethyl phosphonamidic chloride or 2-cyanoethyl, n,n,n,n-tetraisopropyl phosphane are readily available in the market1, and the methods of phosphorylation to produce corresponding phosphoramidites are well known in the art. 1 Manufactured by ChemGenes Corp.